Youth at continuation schools are at higher risk for adult drug abuse problems. Current prevention programs may not fully impact on social influences which participate and maintain drug use in high risk settings wherein users are not a small minority of the teenagers, such as in continuation. Yet, programming has not been implemented and is sorely needed in this settings. The specific aims of this study are: in Year 1, to investigate the impact of normative versus informational social influence on drug use among continuation school students through interviews with staff and students; in Year 1, to compare the immediate impact of standard social influence classroom-based components across comprehensive high school and continuation high school youth; at the end of Year 1 and in Year 2, to develop and pilot a complete social influence curriculum tailored for continuation high school students; in Year 2, to develop a standard community outreach approach to enhance the potential for success of the adapted social influences program; in Year 3, to complete an efficacy trial involving a three-group randomized design with a total of 24 continuation schools, in which the high risk youth-oriented program will be tested with or without a community outreach component (n=8 in each condition), against whatever program these schools usually offer at 8 other continuation schools (control condition), in Years 4 and 5, to complete an intense followup of a subset of these youth to compare the effects of these two programs on use of cigarettes, alcohol, marijuana, and other illicit drugs, as well as measure general changes in prevalence of student drug use at these schools; in Years 4 and 5, as a generalization study, to compare the efficacy of a re-written high-risk youth program with a standard social influences program and a physical consequences attention placebo in a small three-group randomized design involving comprehensive high schools (total n=9 schools); and finally, in Year 4, to test the immediate impact of the modified program at two non-school settings, on a small scale, to explore the application of this program to groups of youth of extreme risk. Taken together, these studies will demonstrate the capacity of social influence concepts to explain and enhance prevention of drug use among higher risk students.